pork-cnn-classification.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "b65dfbdc-bb80-4bd4-8248-dd06483222cd",
   "metadata": {},
   "source": [
    "# Lour's Pork Barrel Classifier (羅老師肉桶文本分類器)🐖\n",
    "## Convolutional Neural Networks for Pork Barrel Project Classification \n",
    "\n",
    "\n",
    "author: davidycliao(David Yen-Chieh Liao)                                                                      \n",
    "email: davidycliao@gmail.com                                                             \n",
    "date: 9-July-2021    "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e8005d5a-f938-4c26-86d5-592408152d08",
   "metadata": {},
   "source": [
    "-------------------------\n",
    "\n",
    "### Stage 1: Libaries & Dependencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "6ca92ba3-7480-418f-91e0-d2beea2d10af",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:38:36.202997Z",
     "iopub.status.busy": "2022-02-22T22:38:36.202490Z",
     "iopub.status.idle": "2022-02-22T22:38:41.035788Z",
     "shell.execute_reply": "2022-02-22T22:38:41.035213Z",
     "shell.execute_reply.started": "2022-02-22T22:38:36.202865Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# built-in library\n",
    "# pip install -r requirements.txt\n",
    "import math\n",
    "import re\n",
    "import collections\n",
    "import zipfile\n",
    "import random\n",
    "from itertools import chain\n",
    "\n",
    "# ML & Deep Learning/ NLP toolkit\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import jieba\n",
    "from sklearn.model_selection import train_test_split\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras import layers\n",
    "import tensorflow_datasets as tfds\n",
    "from tensorflow.keras.callbacks import EarlyStopping,TensorBoard\n",
    "\n",
    "# Visualization\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8bd77e05-355a-422f-87db-2cf77cba3bdd",
   "metadata": {},
   "source": [
    "-------------------------\n",
    "\n",
    "### Stage 2: Data Preprocessing (Training Data: Introduction of Bills and Legislation from 6th Session to 7th Session, 2004-2012)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3b57b9c9-443e-4e26-bda2-15f6877845d5",
   "metadata": {},
   "source": [
    "#### (1) Read file "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "e1973cfb-73d4-4717-a67a-a4e8a0c9edb0",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:38:41.037433Z",
     "iopub.status.busy": "2022-02-22T22:38:41.037275Z",
     "iopub.status.idle": "2022-02-22T22:38:41.083443Z",
     "shell.execute_reply": "2022-02-22T22:38:41.082802Z",
     "shell.execute_reply.started": "2022-02-22T22:38:41.037415Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# read file\n",
    "df = pd.read_csv('data/Pork Bill - 2021-05-20.csv', encoding='utf-8')\n",
    "\n",
    "# combine abstract of bill and title \n",
    "df['text'] =  df['Title'] + df['Content'].fillna(df['Title'])\n",
    "\n",
    "# drop conten without having any characters\n",
    "# view na's row: df[df['text'].isnull()==True]\n",
    "data = df[['text', 'pork_bill']].dropna(subset=['text'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "f6160785-9522-4404-8f77-6df941869cde",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:38:41.084495Z",
     "iopub.status.busy": "2022-02-22T22:38:41.084319Z",
     "iopub.status.idle": "2022-02-22T22:38:41.089334Z",
     "shell.execute_reply": "2022-02-22T22:38:41.088700Z",
     "shell.execute_reply.started": "2022-02-22T22:38:41.084469Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " Pork Legislation 2510 \n",
      " None-Pork Legislation 4733\n"
     ]
    }
   ],
   "source": [
    "print(\" Pork Legislation\", data['pork_bill'].value_counts()[1],'\\n', \n",
    "      \"None-Pork Legislation\", data['pork_bill'].value_counts()[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "11f1b3ab-7f1e-40b0-995a-e66ae58da102",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:38:41.091732Z",
     "iopub.status.busy": "2022-02-22T22:38:41.091510Z",
     "iopub.status.idle": "2022-02-22T22:38:41.217523Z",
     "shell.execute_reply": "2022-02-22T22:38:41.217045Z",
     "shell.execute_reply.started": "2022-02-22T22:38:41.091710Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "plt.figure(figsize=(10, 6))\n",
    "pork_plot = pd.DataFrame({\"num\":data['pork_bill'].value_counts(), \"type\":['Not Pork','Pork']})\n",
    "plt.bar(pork_plot['type'], pork_plot['num'], color='#57A1AB')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5e6d677d-e200-40e6-9252-130eb02fc682",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-16T17:01:03.001989Z",
     "iopub.status.busy": "2021-05-16T17:01:03.001801Z",
     "iopub.status.idle": "2021-05-16T17:01:03.004600Z",
     "shell.execute_reply": "2021-05-16T17:01:03.003789Z",
     "shell.execute_reply.started": "2021-05-16T17:01:03.001969Z"
    }
   },
   "source": [
    "#### (2) Tokenization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "120e8b54-7f75-4ca6-a0d0-9872b1522e8e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:38:41.218414Z",
     "iopub.status.busy": "2022-02-22T22:38:41.218285Z",
     "iopub.status.idle": "2022-02-22T22:38:41.227472Z",
     "shell.execute_reply": "2022-02-22T22:38:41.226462Z",
     "shell.execute_reply.started": "2022-02-22T22:38:41.218397Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import collections\n",
    "import numpy as np\n",
    "import jieba\n",
    "from itertools import chain\n",
    "\n",
    "\n",
    "def jieba_cut(filename):\n",
    "    \"\"\"\n",
    "    cut Chinese and remove stop words\n",
    "    Reference: https://www.cnblogs.com/Luv-GEM/p/10836454.html\n",
    "    Stopwords: https://www.kaggle.com/rikdifos/english-and-chinese-stopwords?select=cn_stopwords.txt\n",
    "    \"\"\"\n",
    "    stop_list = [i.strip() for i in open('data/cn_stopwords.txt','r',encoding='utf-8')]  \n",
    "    news_cut = []\n",
    "    news_list = []\n",
    "    for line in filename:    \n",
    "        if line:\n",
    "            news_cut = list(jieba.cut(''.join(line),cut_all=False,HMM=True))  \n",
    "            news_list.append([word.strip() for word in news_cut if word not in stop_list and len(word.strip())>0]) \n",
    "    news_list = list(chain.from_iterable(news_list))  \n",
    "    return news_list\n",
    "\n",
    "def clearPucts(context):\n",
    "    \"\"\"\n",
    "    remove punctuation\n",
    "    ref: https://chenyuzuoo.github.io/posts/28001/\n",
    "    \"\"\"\n",
    "    context = re.sub(\"[\\s+\\.\\!\\/_,$%^*(+\\\"\\']+|[+——！，。？、~@#￥%……&*（）]+\", \"\", context)\n",
    "    context = re.sub(\"[【】╮╯▽╰╭★→「」]+\",\"\", context)\n",
    "    context = re.sub(\"！，❤。～《》：（）【】「」？”“；：、\",\"\",context)\n",
    "    context = re.sub(\"\\s\",\"\",context)\n",
    "    return context\n",
    "\n",
    "def seg_char(sent):\n",
    "    \"\"\"\n",
    "    cut Chinese and remove stop words\n",
    "    ref: https://blog.csdn.net/renyuanfang/article/details/86487367\n",
    "    \"\"\"\n",
    "    # split\n",
    "    pattern_char_1 = re.compile(r'([\\W])')\n",
    "    parts = pattern_char_1.split(sent)\n",
    "    parts = [p for p in parts if len(p.strip())>0]\n",
    "    # cut sentence\n",
    "    pattern = re.compile(r'([\\u4e00-\\u9fa5])')\n",
    "    chars = pattern.split(sent)\n",
    "    chars = [w for w in chars if len(w.strip())>0]\n",
    "    chars = ' '.join(chars)\n",
    "    return chars"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "db4c6206-6eca-4958-a5d7-6319f48bb348",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:38:41.228540Z",
     "iopub.status.busy": "2022-02-22T22:38:41.228318Z",
     "iopub.status.idle": "2022-02-22T22:38:43.638013Z",
     "shell.execute_reply": "2022-02-22T22:38:43.637420Z",
     "shell.execute_reply.started": "2022-02-22T22:38:41.228517Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "data_clean = [seg_char(text) for text in [clearPucts(text) for text in data.text]]\n",
    "\n",
    "tokenizer = tfds.deprecated.text.SubwordTextEncoder.build_from_corpus(\n",
    "    data_clean, target_vocab_size=2**18)\n",
    "\n",
    "data_inputs = [tokenizer.encode(sentence) for sentence in data_clean]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "089e65af-e49c-4d60-962e-a0a826809aa5",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-16T17:01:47.611294Z",
     "iopub.status.busy": "2021-05-16T17:01:47.611020Z",
     "iopub.status.idle": "2021-05-16T17:01:47.614208Z",
     "shell.execute_reply": "2021-05-16T17:01:47.613487Z",
     "shell.execute_reply.started": "2021-05-16T17:01:47.611276Z"
    }
   },
   "source": [
    "#### (3) Padding"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "b801d231-972c-47e3-83c4-7375c2e11a18",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:38:43.640105Z",
     "iopub.status.busy": "2022-02-22T22:38:43.639876Z",
     "iopub.status.idle": "2022-02-22T22:38:43.691358Z",
     "shell.execute_reply": "2022-02-22T22:38:43.690789Z",
     "shell.execute_reply.started": "2022-02-22T22:38:43.640072Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "MAX_LEN = max([len(sentence) for sentence in data_clean])\n",
    "data_inputs = tf.keras.preprocessing.sequence.pad_sequences(data_inputs,\n",
    "                                                            value=0,\n",
    "                                                            padding=\"post\",\n",
    "                                                            maxlen=MAX_LEN)\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "bd561c2c-9c98-46c3-91a9-9bdb8f160344",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:38:43.692374Z",
     "iopub.status.busy": "2022-02-22T22:38:43.692253Z",
     "iopub.status.idle": "2022-02-22T22:38:43.698248Z",
     "shell.execute_reply": "2022-02-22T22:38:43.697207Z",
     "shell.execute_reply.started": "2022-02-22T22:38:43.692357Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(7243, 785)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_inputs.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f69d90cf-ac3c-4e08-b46b-457d9708d832",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-16T17:02:28.107037Z",
     "iopub.status.busy": "2021-05-16T17:02:28.106843Z",
     "iopub.status.idle": "2021-05-16T17:02:28.109696Z",
     "shell.execute_reply": "2021-05-16T17:02:28.108954Z",
     "shell.execute_reply.started": "2021-05-16T17:02:28.107017Z"
    }
   },
   "source": [
    "#### (4) Spliting Training / Testing Set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "42503355-eea3-4602-ae5a-ea18be84b8d1",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:38:44.943806Z",
     "iopub.status.busy": "2022-02-22T22:38:44.943464Z",
     "iopub.status.idle": "2022-02-22T22:38:44.948000Z",
     "shell.execute_reply": "2022-02-22T22:38:44.946852Z",
     "shell.execute_reply.started": "2022-02-22T22:38:44.943769Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "data_labels = data.pork_bill.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "19235117-f63a-4e06-a102-6efe98c37dbf",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:38:45.467086Z",
     "iopub.status.busy": "2022-02-22T22:38:45.466715Z",
     "iopub.status.idle": "2022-02-22T22:38:45.484003Z",
     "shell.execute_reply": "2022-02-22T22:38:45.483428Z",
     "shell.execute_reply.started": "2022-02-22T22:38:45.467063Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of X Train: (4852, 785) \n",
      "Shape of X Test : (2391, 785) \n",
      "Shape of Y Trian: (4852,) \n",
      "Shape of Y Test : (2391,)\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "from sklearn.model_selection import train_test_split\n",
    "train_inputs, test_inputs, train_labels, test_labels = train_test_split(\n",
    "    data_inputs, data_labels, test_size=0.33, random_state=42)\n",
    "\n",
    "print(\"Shape of X Train:\", train_inputs.shape, '\\n'\n",
    "      \"Shape of X Test :\", test_inputs.shape,'\\n'\n",
    "      \"Shape of Y Trian:\", train_labels.shape , '\\n'\n",
    "      \"Shape of Y Test :\", test_labels.shape )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "6f10cb5c-7c8d-4e1e-a987-11b8de8e42d0",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:38:46.302740Z",
     "iopub.status.busy": "2022-02-22T22:38:46.302506Z",
     "iopub.status.idle": "2022-02-22T22:38:46.306931Z",
     "shell.execute_reply": "2022-02-22T22:38:46.306321Z",
     "shell.execute_reply.started": "2022-02-22T22:38:46.302716Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " Num of Train Set: 4852 \n",
      " Not Pork vs Pork: {0: 3167, 1: 1685}\n"
     ]
    }
   ],
   "source": [
    "unique_train_labels, counts_train_labels = np.unique(train_labels, return_counts=True)\n",
    "\n",
    "print(' Num of Train Set:' , train_inputs.shape[0], '\\n'\n",
    "      ,'Not Pork vs Pork:', dict(zip(unique_train_labels, counts_train_labels)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "f545a34d-f506-4db5-a8df-c0a89cb77a16",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:38:46.801957Z",
     "iopub.status.busy": "2022-02-22T22:38:46.801600Z",
     "iopub.status.idle": "2022-02-22T22:38:46.951539Z",
     "shell.execute_reply": "2022-02-22T22:38:46.950879Z",
     "shell.execute_reply.started": "2022-02-22T22:38:46.801903Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "plt.figure(figsize=(10, 6))\n",
    "train_plot = pd.DataFrame({\"num\":[dict(zip(unique_train_labels, counts_train_labels))[0],\n",
    "                                  dict(zip(unique_train_labels, counts_train_labels))[1]], \n",
    "                           \"type\":['Not Pork','Pork']})\n",
    "plt.bar(train_plot['type'], train_plot['num'], color='#57A1AB')\n",
    "plt.title('Train Set')\n",
    "#plt.show()\n",
    "plt.savefig('images/train_plot.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "fedad623-02ba-4fad-b0b4-a094de7c39de",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:38:47.342643Z",
     "iopub.status.busy": "2022-02-22T22:38:47.342384Z",
     "iopub.status.idle": "2022-02-22T22:38:47.347738Z",
     "shell.execute_reply": "2022-02-22T22:38:47.347056Z",
     "shell.execute_reply.started": "2022-02-22T22:38:47.342624Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " Num of Test Set: 2391 \n",
      " Not Pork vs Pork: {0: 1566, 1: 825}\n"
     ]
    }
   ],
   "source": [
    "unique_test_labels, counts_test_labels = np.unique(test_labels, return_counts=True)\n",
    "\n",
    "print(' Num of Test Set:' , test_inputs.shape[0], '\\n'\n",
    "      ,'Not Pork vs Pork:', dict(zip(unique_test_labels, counts_test_labels)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "ccc6799c-42b8-44c3-8352-e4b68ab3cfee",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:38:47.713614Z",
     "iopub.status.busy": "2022-02-22T22:38:47.713369Z",
     "iopub.status.idle": "2022-02-22T22:38:47.865230Z",
     "shell.execute_reply": "2022-02-22T22:38:47.864522Z",
     "shell.execute_reply.started": "2022-02-22T22:38:47.713577Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "plt.figure(figsize=(10, 6))\n",
    "test_plot = pd.DataFrame({\"num\":[dict(zip(unique_test_labels, counts_test_labels))[0],\n",
    "                                 dict(zip(unique_test_labels, counts_test_labels))[1]], \n",
    "                          \"type\":['Not Pork','Pork']})\n",
    "plt.bar(test_plot['type'], test_plot['num'], color='#B1C5C8')\n",
    "plt.title('Test Set')\n",
    "#plt.show()\n",
    "plt.savefig('images/test_plot.png')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "725f7feb-b93b-4d0b-bcbf-562240a79f8b",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-16T17:04:13.375644Z",
     "iopub.status.busy": "2021-05-16T17:04:13.375440Z",
     "iopub.status.idle": "2021-05-16T17:04:13.378111Z",
     "shell.execute_reply": "2021-05-16T17:04:13.377464Z",
     "shell.execute_reply.started": "2021-05-16T17:04:13.375623Z"
    },
    "tags": []
   },
   "source": [
    "-------------------------\n",
    "\n",
    "\n",
    "### Stage 3: Model and Building"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4a0e74bf-eaf4-4043-8d23-88291d8ea344",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-16T17:13:18.807725Z",
     "iopub.status.busy": "2021-05-16T17:13:18.807498Z",
     "iopub.status.idle": "2021-05-16T17:13:18.810574Z",
     "shell.execute_reply": "2021-05-16T17:13:18.809625Z",
     "shell.execute_reply.started": "2021-05-16T17:13:18.807701Z"
    }
   },
   "source": [
    "#### (1) Using the Subclassing API to Build Dynamic Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "1f18f9b1-5f4d-4d0a-baf4-2eebed7f34f2",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:38:48.988911Z",
     "iopub.status.busy": "2022-02-22T22:38:48.988683Z",
     "iopub.status.idle": "2022-02-22T22:38:48.999364Z",
     "shell.execute_reply": "2022-02-22T22:38:48.998747Z",
     "shell.execute_reply.started": "2022-02-22T22:38:48.988890Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "class DCNN(tf.keras.Model):\n",
    "    def __init__(self,\n",
    "                 vocab_size,\n",
    "                 emb_dim=128,\n",
    "                 nb_filters=100,\n",
    "                 # units: Positive integer, dimensionality of the output space.\n",
    "                 FFN_units=512,\n",
    "                 nb_classes=2,\n",
    "                 dropout_rate=0.1,\n",
    "                 training=False,\n",
    "                 name=\"PorkCNN\"):\n",
    "        super(DCNN, self).__init__(name=name)\n",
    "        self.embedding = layers.Embedding(vocab_size, emb_dim)\n",
    "        self.bigram = layers.Conv1D(filters=nb_filters, kernel_size=2, strides = 1, padding=\"valid\", activation=\"relu\")\n",
    "        self.bigram2 = layers.Conv1D(filters=nb_filters, kernel_size=2, strides = 2, padding=\"valid\", activation=\"relu\")\n",
    "        self.trigram = layers.Conv1D(filters=nb_filters,kernel_size=3,strides = 1, padding=\"valid\",activation=\"relu\")\n",
    "        self.trigram2 = layers.Conv1D(filters=nb_filters,kernel_size=3,strides = 2, padding=\"valid\",activation=\"relu\")\n",
    "        self.fourgram = layers.Conv1D(filters=nb_filters,kernel_size=4,strides = 2, padding=\"valid\",activation=\"relu\")\n",
    "        self.fivegram = layers.Conv1D(filters=nb_filters,kernel_size=5,strides = 2, padding=\"valid\",activation=\"relu\")\n",
    "        self.pool = layers.GlobalMaxPool1D()\n",
    "        self.dense_1 = layers.Dense(units=FFN_units, activation=\"relu\")\n",
    "        self.dropout = layers.Dropout(rate=dropout_rate)\n",
    "        self.last_dense = layers.Dense(units=1, activation=\"sigmoid\")    \n",
    "    def call(self, inputs, training):\n",
    "        x = self.embedding(inputs)\n",
    "        x_1 = self.bigram(x)\n",
    "        x_1 = self.pool(x_1)\n",
    "        x_1_1 = self.bigram2(x)\n",
    "        x_1_1 = self.pool(x_1_1)        \n",
    "        x_2 = self.trigram(x)\n",
    "        x_2 = self.pool(x_2)\n",
    "        x_2_1 = self.trigram2(x)\n",
    "        x_2_1 = self.pool(x_2_1)     \n",
    "        x_3 = self.fourgram(x)\n",
    "        x_3 = self.pool(x_3)\n",
    "        x_4 = self.fourgram(x)\n",
    "        x_4 = self.pool(x_4) \n",
    "        x_5 = self.fivegram(x)\n",
    "        x_5 = self.pool(x_5)      \n",
    "        merged = tf.concat([x_1,x_1_1, x_2,x_2_1, x_3, x_4, x_5], axis=-1) \n",
    "        merged = self.dense_1(merged)\n",
    "        merged = self.dropout(merged, training)\n",
    "        output = self.last_dense(merged)        \n",
    "        return output"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "bfc02e93-3a0e-4c1c-b488-b2db98b53824",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:38:49.706700Z",
     "iopub.status.busy": "2022-02-22T22:38:49.706304Z",
     "iopub.status.idle": "2022-02-22T22:38:49.765080Z",
     "shell.execute_reply": "2022-02-22T22:38:49.764511Z",
     "shell.execute_reply.started": "2022-02-22T22:38:49.706647Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-02-22 22:38:49.726343: I tensorflow/core/platform/cpu_feature_guard.cc:151] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 FMA\n",
      "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n"
     ]
    }
   ],
   "source": [
    "VOCAB_SIZE = tokenizer.vocab_size  #tokenizer.vocab_size  # 5000 tokenizer.vocab_size\n",
    "EMB_DIM = 200\n",
    "NB_FILTERS = 100\n",
    "FFN_UNITS = 256\n",
    "NB_CLASSES = 2 #len(set(train_labels))\n",
    "DROPOUT_RATE = 0.25\n",
    "BATCH_SIZE = 230\n",
    "NB_EPOCHS = 80\n",
    "\n",
    "Dcnn = DCNN(vocab_size=VOCAB_SIZE,\n",
    "            emb_dim=EMB_DIM,\n",
    "            nb_filters=NB_FILTERS,\n",
    "            FFN_units=FFN_UNITS,\n",
    "            nb_classes=NB_CLASSES,\n",
    "            dropout_rate=DROPOUT_RATE)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b0ec8f51-6ed8-49fa-9d5e-c6929184bdf1",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-16T18:43:59.279424Z",
     "iopub.status.busy": "2021-05-16T18:43:59.279202Z",
     "iopub.status.idle": "2021-05-16T18:43:59.282718Z",
     "shell.execute_reply": "2021-05-16T18:43:59.281823Z",
     "shell.execute_reply.started": "2021-05-16T18:43:59.279399Z"
    }
   },
   "source": [
    "#### (2) Compile and Summary of the Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "2cce0414-05eb-4f79-9c27-bca9c91a2903",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:38:53.354379Z",
     "iopub.status.busy": "2022-02-22T22:38:53.354128Z",
     "iopub.status.idle": "2022-02-22T22:38:53.474864Z",
     "shell.execute_reply": "2022-02-22T22:38:53.474289Z",
     "shell.execute_reply.started": "2022-02-22T22:38:53.354359Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"PorkCNN\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " embedding (Embedding)       multiple                  586600    \n",
      "                                                                 \n",
      " conv1d (Conv1D)             multiple                  40100     \n",
      "                                                                 \n",
      " conv1d_1 (Conv1D)           multiple                  40100     \n",
      "                                                                 \n",
      " conv1d_2 (Conv1D)           multiple                  60100     \n",
      "                                                                 \n",
      " conv1d_3 (Conv1D)           multiple                  60100     \n",
      "                                                                 \n",
      " conv1d_4 (Conv1D)           multiple                  80100     \n",
      "                                                                 \n",
      " conv1d_5 (Conv1D)           multiple                  100100    \n",
      "                                                                 \n",
      " global_max_pooling1d (Globa  multiple                 0         \n",
      " lMaxPooling1D)                                                  \n",
      "                                                                 \n",
      " dense (Dense)               multiple                  179456    \n",
      "                                                                 \n",
      " dropout (Dropout)           multiple                  0         \n",
      "                                                                 \n",
      " dense_1 (Dense)             multiple                  257       \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 1,146,913\n",
      "Trainable params: 1,146,913\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "Dcnn.compile(loss=\"binary_crossentropy\", optimizer=\"adam\", metrics=[\"accuracy\"])\n",
    "Dcnn.build(input_shape = (train_inputs.shape[1], EMB_DIM\n",
    "                         )) # (train_inputs.shape[1] , EMB_DIM)  (785 , EMB_DIM)\n",
    "Dcnn.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3880d2a4-2152-43de-bad3-34e67e97469e",
   "metadata": {
    "tags": []
   },
   "source": [
    "#### (3) Check Point Path"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "2c18aa59-a797-4a67-bc4d-a412dbaaedcb",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:38:54.237314Z",
     "iopub.status.busy": "2022-02-22T22:38:54.237003Z",
     "iopub.status.idle": "2022-02-22T22:38:54.244765Z",
     "shell.execute_reply": "2022-02-22T22:38:54.244126Z",
     "shell.execute_reply.started": "2022-02-22T22:38:54.237261Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "checkpoint_path = \"checkpoint_recode/ckpt_manager\"\n",
    "\n",
    "ckpt = tf.train.Checkpoint(Dcnn=Dcnn)\n",
    "\n",
    "ckpt_manager = tf.train.CheckpointManager(ckpt, checkpoint_path, max_to_keep=10)\n",
    "\n",
    "if ckpt_manager.latest_checkpoint:\n",
    "    ckpt.restore(ckpt_manager.latest_checkpoint)\n",
    "    print(\"Checkpoint Located!!\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "73d70667-275f-40c0-bd71-6b47e9722798",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-16T17:47:13.442664Z",
     "iopub.status.busy": "2021-05-16T17:47:13.442381Z",
     "iopub.status.idle": "2021-05-16T17:47:13.448371Z",
     "shell.execute_reply": "2021-05-16T17:47:13.446970Z",
     "shell.execute_reply.started": "2021-05-16T17:47:13.442620Z"
    }
   },
   "source": [
    "-------------------------\n",
    "\n",
    "### Stage 4: Train the Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "cc975047-5a58-46dc-b427-262796d9b625",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:38:55.214491Z",
     "iopub.status.busy": "2022-02-22T22:38:55.214153Z",
     "iopub.status.idle": "2022-02-22T22:38:55.218905Z",
     "shell.execute_reply": "2022-02-22T22:38:55.218002Z",
     "shell.execute_reply.started": "2022-02-22T22:38:55.214467Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "early_stop = EarlyStopping(monitor='val_loss', patience=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "baf0cade-2c45-43b2-9782-6a8c6e570153",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:38:58.179932Z",
     "iopub.status.busy": "2022-02-22T22:38:58.179582Z",
     "iopub.status.idle": "2022-02-22T22:46:04.490133Z",
     "shell.execute_reply": "2022-02-22T22:46:04.489264Z",
     "shell.execute_reply.started": "2022-02-22T22:38:58.179909Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/80\n",
      "22/22 [==============================] - 40s 2s/step - loss: 0.6059 - accuracy: 0.6494 - val_loss: 0.5103 - val_accuracy: 0.7808\n",
      "Epoch 2/80\n",
      "22/22 [==============================] - 42s 2s/step - loss: 0.3754 - accuracy: 0.8528 - val_loss: 0.2575 - val_accuracy: 0.8950\n",
      "Epoch 3/80\n",
      "22/22 [==============================] - 47s 2s/step - loss: 0.1758 - accuracy: 0.9334 - val_loss: 0.1780 - val_accuracy: 0.9318\n",
      "Epoch 4/80\n",
      "22/22 [==============================] - 50s 2s/step - loss: 0.0890 - accuracy: 0.9683 - val_loss: 0.1537 - val_accuracy: 0.9452\n",
      "Epoch 5/80\n",
      "22/22 [==============================] - 53s 2s/step - loss: 0.0518 - accuracy: 0.9835 - val_loss: 0.1697 - val_accuracy: 0.9444\n",
      "Epoch 6/80\n",
      "22/22 [==============================] - 46s 2s/step - loss: 0.0610 - accuracy: 0.9819 - val_loss: 0.1987 - val_accuracy: 0.9335\n",
      "Epoch 7/80\n",
      "22/22 [==============================] - 47s 2s/step - loss: 0.0559 - accuracy: 0.9860 - val_loss: 0.1735 - val_accuracy: 0.9452\n",
      "Epoch 8/80\n",
      "22/22 [==============================] - 55s 3s/step - loss: 0.0411 - accuracy: 0.9889 - val_loss: 0.1817 - val_accuracy: 0.9440\n",
      "Epoch 9/80\n",
      "22/22 [==============================] - 44s 2s/step - loss: 0.0412 - accuracy: 0.9868 - val_loss: 0.1859 - val_accuracy: 0.9406\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'checkpoint_recode/ckpt_manager/ckpt-1'"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Dcnn.fit(train_inputs, train_labels,\n",
    "         validation_data=(test_inputs, test_labels),\n",
    "         batch_size=BATCH_SIZE,\n",
    "         epochs=NB_EPOCHS,\n",
    "         callbacks=[early_stop])\n",
    "ckpt_manager.save()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3b0ac256-ab81-4ff5-8430-48f8bba50126",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-16T18:23:25.926977Z",
     "iopub.status.busy": "2021-05-16T18:23:25.926754Z",
     "iopub.status.idle": "2021-05-16T18:23:25.931072Z",
     "shell.execute_reply": "2021-05-16T18:23:25.930041Z",
     "shell.execute_reply.started": "2021-05-16T18:23:25.926952Z"
    }
   },
   "source": [
    "-------------------------\n",
    "\n",
    "### Stage 5: Evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "58099cc1-19e0-443a-b960-7ae4d67344d3",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:46:04.492705Z",
     "iopub.status.busy": "2022-02-22T22:46:04.492383Z",
     "iopub.status.idle": "2022-02-22T22:46:04.516373Z",
     "shell.execute_reply": "2022-02-22T22:46:04.515459Z",
     "shell.execute_reply.started": "2022-02-22T22:46:04.492669Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>loss</th>\n",
       "      <th>accuracy</th>\n",
       "      <th>val_loss</th>\n",
       "      <th>val_accuracy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.605947</td>\n",
       "      <td>0.649423</td>\n",
       "      <td>0.510334</td>\n",
       "      <td>0.780845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.375364</td>\n",
       "      <td>0.852844</td>\n",
       "      <td>0.257516</td>\n",
       "      <td>0.895023</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.175829</td>\n",
       "      <td>0.933430</td>\n",
       "      <td>0.178023</td>\n",
       "      <td>0.931828</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.088962</td>\n",
       "      <td>0.968261</td>\n",
       "      <td>0.153718</td>\n",
       "      <td>0.945211</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.051790</td>\n",
       "      <td>0.983512</td>\n",
       "      <td>0.169675</td>\n",
       "      <td>0.944375</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.061041</td>\n",
       "      <td>0.981863</td>\n",
       "      <td>0.198732</td>\n",
       "      <td>0.933501</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.055922</td>\n",
       "      <td>0.985985</td>\n",
       "      <td>0.173460</td>\n",
       "      <td>0.945211</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.041117</td>\n",
       "      <td>0.988871</td>\n",
       "      <td>0.181670</td>\n",
       "      <td>0.943956</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.041197</td>\n",
       "      <td>0.986810</td>\n",
       "      <td>0.185922</td>\n",
       "      <td>0.940611</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       loss  accuracy  val_loss  val_accuracy\n",
       "0  0.605947  0.649423  0.510334      0.780845\n",
       "1  0.375364  0.852844  0.257516      0.895023\n",
       "2  0.175829  0.933430  0.178023      0.931828\n",
       "3  0.088962  0.968261  0.153718      0.945211\n",
       "4  0.051790  0.983512  0.169675      0.944375\n",
       "5  0.061041  0.981863  0.198732      0.933501\n",
       "6  0.055922  0.985985  0.173460      0.945211\n",
       "7  0.041117  0.988871  0.181670      0.943956\n",
       "8  0.041197  0.986810  0.185922      0.940611"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "losses = pd.DataFrame(Dcnn.history.history)\n",
    "losses"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "b1b62bf7-433e-4a9e-b0ed-a6b57515b304",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:46:04.518302Z",
     "iopub.status.busy": "2022-02-22T22:46:04.517920Z",
     "iopub.status.idle": "2022-02-22T22:46:04.975181Z",
     "shell.execute_reply": "2022-02-22T22:46:04.974273Z",
     "shell.execute_reply.started": "2022-02-22T22:46:04.518259Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "losses.plot(figsize=(8,5))\n",
    "plt.grid(True)\n",
    "plt.gca().set_ylim(0,1)\n",
    "plt.xlabel(\"Learning Curves: the mean training loss and accuracy measured over each epochs\")\n",
    "plt.savefig('images/learning_curves3.png')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "17760825-eb35-4f8c-88a0-45e94387c08d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:46:04.978148Z",
     "iopub.status.busy": "2022-02-22T22:46:04.977805Z",
     "iopub.status.idle": "2022-02-22T22:46:09.726812Z",
     "shell.execute_reply": "2022-02-22T22:46:09.725629Z",
     "shell.execute_reply.started": "2022-02-22T22:46:04.978111Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "11/11 [==============================] - 5s 421ms/step - loss: 0.1859 - accuracy: 0.9406\n",
      "[0.18592208623886108, 0.9406106472015381]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report,confusion_matrix\n",
    "evaluation_model = Dcnn.evaluate(test_inputs, test_labels, batch_size=BATCH_SIZE)\n",
    "print(evaluation_model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "c9a3f93d-4398-418c-b511-69594874cc5a",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:47:20.909353Z",
     "iopub.status.busy": "2022-02-22T22:47:20.908943Z",
     "iopub.status.idle": "2022-02-22T22:47:23.756292Z",
     "shell.execute_reply": "2022-02-22T22:47:23.755731Z",
     "shell.execute_reply.started": "2022-02-22T22:47:20.909327Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.95      0.97      0.96      1566\n",
      "           1       0.93      0.89      0.91       825\n",
      "\n",
      "    accuracy                           0.94      2391\n",
      "   macro avg       0.94      0.93      0.94      2391\n",
      "weighted avg       0.94      0.94      0.94      2391\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# pd.DataFrame(confusion_matrix(test_labels,predictions))\n",
    "predictions = Dcnn.predict(test_inputs)\n",
    "predictions = np.where(predictions >0.8 , 1, 0)\n",
    "print(classification_report(test_labels, predictions))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "4767688b-bfc8-4dc4-b8ec-2252f7791534",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:47:24.961621Z",
     "iopub.status.busy": "2022-02-22T22:47:24.961198Z",
     "iopub.status.idle": "2022-02-22T22:47:24.973247Z",
     "shell.execute_reply": "2022-02-22T22:47:24.972720Z",
     "shell.execute_reply.started": "2022-02-22T22:47:24.961576Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Predictions: Not Pork(0)</th>\n",
       "      <th>Predictions:Pork(1)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Acutal: Not Pork(0)</th>\n",
       "      <td>1514</td>\n",
       "      <td>52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Acutal: Pork (1)</th>\n",
       "      <td>87</td>\n",
       "      <td>738</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     Predictions: Not Pork(0)  Predictions:Pork(1)\n",
       "Acutal: Not Pork(0)                      1514                   52\n",
       "Acutal: Pork (1)                           87                  738"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t = pd.DataFrame(confusion_matrix(test_labels,predictions), \n",
    "             columns=['Predictions: Not Pork(0)','Predictions:Pork(1)'])\n",
    "t.index = ['Acutal: Not Pork(0)', 'Acutal: Pork (1)']\n",
    "t"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d0416d6f-5e46-47c9-abad-f344bea7a94a",
   "metadata": {},
   "source": [
    "-------------------------\n",
    "\n",
    "### Stage 6: Try the Model with New Data and Export an End-to-end Model\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "394b9a6e-0130-4af2-88e5-efe91cf9f348",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-16T22:53:41.558105Z",
     "iopub.status.busy": "2021-05-16T22:53:41.557872Z",
     "iopub.status.idle": "2021-05-16T22:53:41.560984Z",
     "shell.execute_reply": "2021-05-16T22:53:41.560122Z",
     "shell.execute_reply.started": "2021-05-16T22:53:41.558080Z"
    }
   },
   "source": [
    "#### (1) Test New Dataset from 2000 samples from Legislative Questions from 6th Session "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "632bf33e-1df7-46c5-9a67-ab7141a949c2",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:47:26.690567Z",
     "iopub.status.busy": "2022-02-22T22:47:26.690372Z",
     "iopub.status.idle": "2022-02-22T22:47:26.693986Z",
     "shell.execute_reply": "2022-02-22T22:47:26.693201Z",
     "shell.execute_reply.started": "2022-02-22T22:47:26.690547Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def as_num(x):\n",
    "    \"\"\"\n",
    "    keep 10 decimals\n",
    "    \"\"\"\n",
    "    y = '{:.15f}'.format(x) \n",
    "    return y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "06ea7d54-49e9-4363-8304-d1dabaf3beb4",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:47:26.882469Z",
     "iopub.status.busy": "2022-02-22T22:47:26.882221Z",
     "iopub.status.idle": "2022-02-22T22:47:27.166975Z",
     "shell.execute_reply": "2022-02-22T22:47:27.166372Z",
     "shell.execute_reply.started": "2022-02-22T22:47:26.882450Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import random\n",
    "\n",
    "# test the trined model using new text from 200 samples from Leislative Questions\n",
    "LQ6 = pd.read_csv('data/LQ_6th.csv')\n",
    "\n",
    "# drop '0': legislor  never have proposed any legisaltive questions. \n",
    "LQ6 = LQ6[LQ6['title'].isin(['0'])== False]\n",
    "\n",
    "# select text with number of charecters within EMB_DIM; len(sub_set) = 11162\n",
    "# sample_df = LQ6.loc[[len(char_num) for char_num in LQ6.title if len(char_num) < MAX_LEN], ['legislator','title', 'topic', 'category', 'keywords', 'ques_type']].reset_index(drop=True)\n",
    "random.seed(42)\n",
    "sample_df = LQ6.iloc[random.sample(range(len(LQ6)), 2000)][['legislator', 'title', 'topic', 'category', 'keywords', 'ques_type']]\n",
    "sub_set = [seg_char(text) for text in [clearPucts(text) for text in sample_df.title]]\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "68cc06a7-27d7-4007-949d-f7df81246e25",
   "metadata": {
    "tags": []
   },
   "source": [
    "##### Top 10 of 2000 Samples"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "d4b618c9-8c4b-40ec-a396-b4679233197e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:47:27.181128Z",
     "iopub.status.busy": "2022-02-22T22:47:27.180887Z",
     "iopub.status.idle": "2022-02-22T22:47:41.767231Z",
     "shell.execute_reply": "2022-02-22T22:47:41.766540Z",
     "shell.execute_reply.started": "2022-02-22T22:47:27.181063Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "top = pd.DataFrame(\n",
    "    {'legislator': sample_df.legislator,\n",
    "     'Pork Value(Constituency Interest)':[as_num(Dcnn(np.array([tokenizer.encode(line)]), training=True).numpy()[0][0]) for line in sub_set],\n",
    "     'Legislative Questions ': sample_df.title,\n",
    "     'Topic': sample_df.topic,\n",
    "     'Key Word':sample_df.keywords}).sort_values(by=['Pork Value(Constituency Interest)'], ascending = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "ebb1af00-93a6-4512-917c-3a817dcb1ba3",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:47:41.768799Z",
     "iopub.status.busy": "2022-02-22T22:47:41.768629Z",
     "iopub.status.idle": "2022-02-22T22:47:41.781526Z",
     "shell.execute_reply": "2022-02-22T22:47:41.780729Z",
     "shell.execute_reply.started": "2022-02-22T22:47:41.768780Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>legislator</th>\n",
       "      <th>Pork Value(Constituency Interest)</th>\n",
       "      <th>Legislative Questions</th>\n",
       "      <th>Topic</th>\n",
       "      <th>Key Word</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9892</th>\n",
       "      <td>陳啟昱</td>\n",
       "      <td>0.998417615890503</td>\n",
       "      <td>鑑於現行《所得稅法》第十七條規定特別扣除額教育支出部分，僅以納稅義務人之子女就讀大專院校為限...</td>\n",
       "      <td>所得稅 ; 教育費用 ; 扣除額</td>\n",
       "      <td>所得稅法 ; 特別扣除額 ; 教育支出</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3999</th>\n",
       "      <td>林正峰</td>\n",
       "      <td>0.997364819049835</td>\n",
       "      <td>針對政府準備修法推動「二代健保」，健保保費採取「年度所得總額」為計算基礎，而非採用扣除免稅額...</td>\n",
       "      <td>國民 ; 健康保險 ; 保險費</td>\n",
       "      <td>二代健保 ; 年度所得總額</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1761</th>\n",
       "      <td>彭添富</td>\n",
       "      <td>0.996990680694580</td>\n",
       "      <td>針對「辦理九十四年原住民中低收入戶家庭租屋補助計畫」專案補助計畫，特向行政院提出質詢。</td>\n",
       "      <td>原住民族生活</td>\n",
       "      <td>原住民中低收入戶 ; 租屋補助</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>丁守中</td>\n",
       "      <td>0.994604468345642</td>\n",
       "      <td>針就民眾陳情指出，目前政府對身心障礙者提供之生活津貼，依身心障礙程度等級分為1000元至50...</td>\n",
       "      <td>身心障礙者福利</td>\n",
       "      <td>身心障礙者生活津貼</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1706</th>\n",
       "      <td>彭添富</td>\n",
       "      <td>0.993997812271118</td>\n",
       "      <td>針對「觀音鄉保生社區風貌營造規劃設計」專案補助計畫，特向行政院提出質詢。</td>\n",
       "      <td>社區計畫 ; 政府補助</td>\n",
       "      <td>保生社區</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>188</th>\n",
       "      <td>丁守中</td>\n",
       "      <td>0.992515802383423</td>\n",
       "      <td>針就「國軍退除役官兵就養安置辦法」第七條第一項第五款規定：「全家人口所有之土地及房屋公告現值...</td>\n",
       "      <td>退除役官兵 ; 榮民生活輔導</td>\n",
       "      <td>國軍退除役官兵就養安置辦法</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>619</th>\n",
       "      <td>吳成典</td>\n",
       "      <td>0.990735411643982</td>\n",
       "      <td>鑑於居家安全及消防常識已逐漸受到民眾的重視，請中央寬列消防機關人事費用，補助金、馬兩縣規劃合...</td>\n",
       "      <td>住宅安全 ; 消防設施</td>\n",
       "      <td>居家安全 ; 消防常識</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1641</th>\n",
       "      <td>彭添富</td>\n",
       "      <td>0.988783240318298</td>\n",
       "      <td>針對「豪雨成災，農作物損失補償」問題，特向行政院提出質詢。</td>\n",
       "      <td>農業補助</td>\n",
       "      <td>豪雨 ; 農作物</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8340</th>\n",
       "      <td>蔡煌瑯(離職)</td>\n",
       "      <td>0.986971139907837</td>\n",
       "      <td>「2005南投花卉嘉年華」活動爭取中央經費補助案由，特向行政院提出質詢。</td>\n",
       "      <td>農業推廣 ; 政府補助</td>\n",
       "      <td>南投花卉嘉年華 ; 中央經費</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9527</th>\n",
       "      <td>鄭朝明</td>\n",
       "      <td>0.986645400524139</td>\n",
       "      <td>針對我國生育率逐年下降，嚴重影響人口結構發展。民眾不願多生育子女，主因在於負擔費用過重，政府...</td>\n",
       "      <td>出生率 ; 人口政策</td>\n",
       "      <td>生育率 ; 人口 ; 人口老化</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     legislator Pork Value(Constituency Interest)  \\\n",
       "9892        陳啟昱                 0.998417615890503   \n",
       "3999        林正峰                 0.997364819049835   \n",
       "1761        彭添富                 0.996990680694580   \n",
       "35          丁守中                 0.994604468345642   \n",
       "1706        彭添富                 0.993997812271118   \n",
       "188         丁守中                 0.992515802383423   \n",
       "619         吳成典                 0.990735411643982   \n",
       "1641        彭添富                 0.988783240318298   \n",
       "8340    蔡煌瑯(離職)                 0.986971139907837   \n",
       "9527        鄭朝明                 0.986645400524139   \n",
       "\n",
       "                                 Legislative Questions              Topic  \\\n",
       "9892  鑑於現行《所得稅法》第十七條規定特別扣除額教育支出部分，僅以納稅義務人之子女就讀大專院校為限...  所得稅 ; 教育費用 ; 扣除額   \n",
       "3999  針對政府準備修法推動「二代健保」，健保保費採取「年度所得總額」為計算基礎，而非採用扣除免稅額...   國民 ; 健康保險 ; 保險費   \n",
       "1761        針對「辦理九十四年原住民中低收入戶家庭租屋補助計畫」專案補助計畫，特向行政院提出質詢。            原住民族生活   \n",
       "35    針就民眾陳情指出，目前政府對身心障礙者提供之生活津貼，依身心障礙程度等級分為1000元至50...           身心障礙者福利   \n",
       "1706               針對「觀音鄉保生社區風貌營造規劃設計」專案補助計畫，特向行政院提出質詢。       社區計畫 ; 政府補助   \n",
       "188   針就「國軍退除役官兵就養安置辦法」第七條第一項第五款規定：「全家人口所有之土地及房屋公告現值...    退除役官兵 ; 榮民生活輔導   \n",
       "619   鑑於居家安全及消防常識已逐漸受到民眾的重視，請中央寬列消防機關人事費用，補助金、馬兩縣規劃合...       住宅安全 ; 消防設施   \n",
       "1641                      針對「豪雨成災，農作物損失補償」問題，特向行政院提出質詢。              農業補助   \n",
       "8340               「2005南投花卉嘉年華」活動爭取中央經費補助案由，特向行政院提出質詢。       農業推廣 ; 政府補助   \n",
       "9527  針對我國生育率逐年下降，嚴重影響人口結構發展。民眾不願多生育子女，主因在於負擔費用過重，政府...        出生率 ; 人口政策   \n",
       "\n",
       "                 Key Word  \n",
       "9892  所得稅法 ; 特別扣除額 ; 教育支出  \n",
       "3999        二代健保 ; 年度所得總額  \n",
       "1761      原住民中低收入戶 ; 租屋補助  \n",
       "35              身心障礙者生活津貼  \n",
       "1706                 保生社區  \n",
       "188         國軍退除役官兵就養安置辦法  \n",
       "619           居家安全 ; 消防常識  \n",
       "1641             豪雨 ; 農作物  \n",
       "8340       南投花卉嘉年華 ; 中央經費  \n",
       "9527      生育率 ; 人口 ; 人口老化  "
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "top.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6a6d4b15-3e05-427d-a664-8c8ce4e6622c",
   "metadata": {},
   "source": [
    "##### Last 10 Rows of 2000 Samples"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "5444905f-b027-4116-9ecc-1e9dec58dd45",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:47:41.783432Z",
     "iopub.status.busy": "2022-02-22T22:47:41.782858Z",
     "iopub.status.idle": "2022-02-22T22:47:41.793614Z",
     "shell.execute_reply": "2022-02-22T22:47:41.793050Z",
     "shell.execute_reply.started": "2022-02-22T22:47:41.783406Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>legislator</th>\n",
       "      <th>Pork Value(Constituency Interest)</th>\n",
       "      <th>Legislative Questions</th>\n",
       "      <th>Topic</th>\n",
       "      <th>Key Word</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>10120</th>\n",
       "      <td>陳朝龍</td>\n",
       "      <td>0.000003376766927</td>\n",
       "      <td>針對英國政府宣稱台灣出口至該國禽鳥，檢驗出感染禽流感H5N1病毒死亡。由於我國迄今並未發現有...</td>\n",
       "      <td>傳染病防治 ; 走私</td>\n",
       "      <td>英國政府 ; 台灣禽鳥 ; 禽流感 ; 相思鳥 ; 走私</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6347</th>\n",
       "      <td>王幸男</td>\n",
       "      <td>0.000003163653446</td>\n",
       "      <td>針對本院交通委員會在台中市長夫人邵曉鈴女士車禍事件之後，以超高的效率通過了「道路交通管理處罰...</td>\n",
       "      <td>交通規則</td>\n",
       "      <td>道路交通管理處罰條例 ; 安全帶</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2866</th>\n",
       "      <td>李鎮楠</td>\n",
       "      <td>0.000002799774620</td>\n",
       "      <td>針對國內近來頻傳黑心食品充斥市面，引起消費者恐慌，提出相關質疑。黑心事件層出不窮，餵豬澱粉回...</td>\n",
       "      <td>食品管理</td>\n",
       "      <td>黑心食品 ; 食品認證</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2439</th>\n",
       "      <td>李俊毅</td>\n",
       "      <td>0.000002592328201</td>\n",
       "      <td>針對近來國務機要費與首長特別費因相關規範未臻明確，遭質疑該筆金額等同於首長薪水或私房錢，且有...</td>\n",
       "      <td>國家元首 ; 行政首長 ; 統一發票</td>\n",
       "      <td>國務機要費 ; 首長特別費</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018</th>\n",
       "      <td>曹來旺</td>\n",
       "      <td>0.000002244937150</td>\n",
       "      <td>針對政府宣示廢除死刑多年，雖然死刑的法律仍在，但判處死刑的案例減少，即使判了死刑，也常拖著不...</td>\n",
       "      <td>死刑</td>\n",
       "      <td>死刑 ; 人權</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6532</th>\n",
       "      <td>王幸男</td>\n",
       "      <td>0.000002108804438</td>\n",
       "      <td>針對陳總統日昨重申，有信心一定能夠在二○○八年卸任總統之前，為台灣催生一部合身、合時、合用的...</td>\n",
       "      <td>憲法修正</td>\n",
       "      <td>新憲法 ; 修改憲法</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8264</th>\n",
       "      <td>蔡其昌</td>\n",
       "      <td>0.000001428694077</td>\n",
       "      <td>針對近年來詐騙集團盛行，傳送大量簡訊詐財，雖內政部警政署自八月一日起，特別提供簡訊特碼「09...</td>\n",
       "      <td>詐欺 ; 電話</td>\n",
       "      <td>詐騙 ; 來電顯號</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7514</th>\n",
       "      <td>紀國棟</td>\n",
       "      <td>0.000001150604248</td>\n",
       "      <td>鑒於台灣極力發展觀光倍增之際，竟一再傳出中國觀光客來台糾紛頻傳，爰要求觀光局為下列改進：一、...</td>\n",
       "      <td>旅遊業 ; 觀光</td>\n",
       "      <td>中國觀光客 ; 旅遊糾紛</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>982</th>\n",
       "      <td>周守訓</td>\n",
       "      <td>0.000001047020760</td>\n",
       "      <td>針對日前媒體報導陳水扁總統宣示，十二月二十五日耶誕節是重要的宗教節慶，更是我國的行憲紀念日，...</td>\n",
       "      <td>宗教 ; 假日</td>\n",
       "      <td>耶誕節 ; 行憲紀念日 ; 宗教自由 ; 政教分離</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6636</th>\n",
       "      <td>王幸男</td>\n",
       "      <td>0.000000255493234</td>\n",
       "      <td>針對「莊嚴肅穆」的行憲紀念日，在威權時代「今天應懸掛國旗」，不過，在愈來愈多人的真實生活中，...</td>\n",
       "      <td>紀念活動 ; 憲政運作</td>\n",
       "      <td>聖誕節 ; 行憲紀念日 ; 制憲權</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      legislator Pork Value(Constituency Interest)  \\\n",
       "10120        陳朝龍                 0.000003376766927   \n",
       "6347         王幸男                 0.000003163653446   \n",
       "2866         李鎮楠                 0.000002799774620   \n",
       "2439         李俊毅                 0.000002592328201   \n",
       "2018         曹來旺                 0.000002244937150   \n",
       "6532         王幸男                 0.000002108804438   \n",
       "8264         蔡其昌                 0.000001428694077   \n",
       "7514         紀國棟                 0.000001150604248   \n",
       "982          周守訓                 0.000001047020760   \n",
       "6636         王幸男                 0.000000255493234   \n",
       "\n",
       "                                  Legislative Questions                Topic  \\\n",
       "10120  針對英國政府宣稱台灣出口至該國禽鳥，檢驗出感染禽流感H5N1病毒死亡。由於我國迄今並未發現有...          傳染病防治 ; 走私   \n",
       "6347   針對本院交通委員會在台中市長夫人邵曉鈴女士車禍事件之後，以超高的效率通過了「道路交通管理處罰...                交通規則   \n",
       "2866   針對國內近來頻傳黑心食品充斥市面，引起消費者恐慌，提出相關質疑。黑心事件層出不窮，餵豬澱粉回...                食品管理   \n",
       "2439   針對近來國務機要費與首長特別費因相關規範未臻明確，遭質疑該筆金額等同於首長薪水或私房錢，且有...  國家元首 ; 行政首長 ; 統一發票   \n",
       "2018   針對政府宣示廢除死刑多年，雖然死刑的法律仍在，但判處死刑的案例減少，即使判了死刑，也常拖著不...                  死刑   \n",
       "6532   針對陳總統日昨重申，有信心一定能夠在二○○八年卸任總統之前，為台灣催生一部合身、合時、合用的...                憲法修正   \n",
       "8264   針對近年來詐騙集團盛行，傳送大量簡訊詐財，雖內政部警政署自八月一日起，特別提供簡訊特碼「09...             詐欺 ; 電話   \n",
       "7514   鑒於台灣極力發展觀光倍增之際，竟一再傳出中國觀光客來台糾紛頻傳，爰要求觀光局為下列改進：一、...            旅遊業 ; 觀光   \n",
       "982    針對日前媒體報導陳水扁總統宣示，十二月二十五日耶誕節是重要的宗教節慶，更是我國的行憲紀念日，...             宗教 ; 假日   \n",
       "6636   針對「莊嚴肅穆」的行憲紀念日，在威權時代「今天應懸掛國旗」，不過，在愈來愈多人的真實生活中，...         紀念活動 ; 憲政運作   \n",
       "\n",
       "                           Key Word  \n",
       "10120  英國政府 ; 台灣禽鳥 ; 禽流感 ; 相思鳥 ; 走私  \n",
       "6347               道路交通管理處罰條例 ; 安全帶  \n",
       "2866                    黑心食品 ; 食品認證  \n",
       "2439                  國務機要費 ; 首長特別費  \n",
       "2018                        死刑 ; 人權  \n",
       "6532                     新憲法 ; 修改憲法  \n",
       "8264                      詐騙 ; 來電顯號  \n",
       "7514                   中國觀光客 ; 旅遊糾紛  \n",
       "982       耶誕節 ; 行憲紀念日 ; 宗教自由 ; 政教分離  \n",
       "6636              聖誕節 ; 行憲紀念日 ; 制憲權  "
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "top.tail(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5dc01475-847a-408f-9972-ac02e7aa4721",
   "metadata": {},
   "source": [
    "#### (2) Export an End-to-end Model\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "196028d0-edfe-434f-a173-4ccaa688faaa",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:47:41.795883Z",
     "iopub.status.busy": "2022-02-22T22:47:41.795664Z",
     "iopub.status.idle": "2022-02-22T22:47:41.799851Z",
     "shell.execute_reply": "2022-02-22T22:47:41.799237Z",
     "shell.execute_reply.started": "2022-02-22T22:47:41.795858Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "from tensorflow.keras.models import load_model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "071291d0-c5bb-48de-adb2-f8b2bb3bb542",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-02-22T22:47:41.800872Z",
     "iopub.status.busy": "2022-02-22T22:47:41.800680Z",
     "iopub.status.idle": "2022-02-22T22:47:42.922258Z",
     "shell.execute_reply": "2022-02-22T22:47:42.921750Z",
     "shell.execute_reply.started": "2022-02-22T22:47:41.800849Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-02-22 22:47:42.366664: W tensorflow/python/util/util.cc:368] Sets are not currently considered sequences, but this may change in the future, so consider avoiding using them.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: lour_pork_model/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: lour_pork_model/assets\n"
     ]
    }
   ],
   "source": [
    "# Get model (Sequential, Functional Model, or Model subclass)\n",
    "Dcnn.save('lour_pork_model')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2ac46183-b841-469b-8897-1fea2922e445",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from tensorflow import keras\n",
    "model = keras.models.load_model('lour_pork_model')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.11"
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "state": {},
    "version_major": 2,
    "version_minor": 0
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}